What is Memex and its primary purpose?

Memex is an open-source tool that acts as a local knowledge base, providing AI agents with persistent memory for technical guides and reusable contexts to enhance their intelligence and consistency.

How does Memex integrate with AI models like Claude?

Memex integrates seamlessly with Claude 3.7+ via the Model Context Protocol (MCP), allowing AI agents to directly create, update, search, and delete content within the knowledge base.

What types of information can Memex store?

It manages a dual knowledge base: 'guides' for step-by-step technical instructions and 'contexts' for AI personas, conventions, or specific prompts, all powered by semantic search.

What technology powers Memex's search capabilities?

Memex uses semantic vector search powered by Ollama embeddings (specifically 'nomic-embed-text') to enable intelligent and highly relevant retrieval of information from your knowledge base.

Memex

Name: Memex
Author: jacquesbh

byjacquesbh

•

Developer Tools

Learning & Documentation

Productivity & Workflow

Manages a local knowledge base of technical guides and reusable contexts for AI agents, enabling persistent access to specialized information.

Inspired by Vannevar Bush's visionary 1945 concept, Memex operates as a local Model Context Protocol (MCP) server dedicated to augmenting AI memory. It provides persistent access to a comprehensive knowledge base, distinguishing between step-by-step technical guides and reusable contexts (such as personas, conventions, and prompts). This empowers AI agents with structured, semantically searchable information, fostering enhanced interactions and more effective task execution within development and operational workflows.

Key Features

01Dual knowledge base for technical guides and AI interaction contexts

02Tools for creating, updating, searching, listing, and deleting content directly from AI agents

03Semantic vector search powered by Ollama embeddings for intelligent retrieval

04Seamless integration with Claude 3.7+ via the Model Context Protocol

05Flexible storage for knowledge bases, configurable for project-specific or global use

063 GitHub stars

Use Cases

01Providing AI agents with persistent, searchable access to technical documentation and how-to guides.

02Supplying AI models with reusable personas, conventions, and prompt templates for consistent interactions.

03Augmenting AI memory for complex tasks requiring instant recall of specialized information or contextual data.